Secondary Logo

Journal Logo

The Use of High-Fidelity Human Patient Simulation and the Introduction of New Anesthesia Delivery Systems

Dalley, Paul MbChB*; Robinson, Brian PhD; Weller, Jennifer MClinEd, FANZCA; Caldwell, Catherine MbBCh, FANZCA*

doi: 10.1213/01.ANE.0000136804.46675.EA
Technology, Computing, and Simulation: Research Report
Free
SDC

New anesthesia delivery systems are becoming increasingly complex. Although equipment is involved in a large proportion of intraoperative anesthesia problems (most also involving human error), the current methods of introducing new equipment into clinical practice have not been well studied. We designed a randomized, controlled, prospective study to investigate an alternative method of introducing new anesthesia equipment. Fifteen anesthesiology trainees were randomized to either the standard introduction to a Dräger Fabius GS anesthesia delivery machine plus simulated clinical use of the new machine in a high-fidelity human patient simulator (HPS) (Group 1) or to the standard introduction alone (Group 2). We used a questionnaire to seek their opinion on the new equipment, and responses showed that both groups were comparable in their reported confidence to use the new equipment safely. All trainees were then tested in two simulated anesthetic crises with the new machine. Performance was analyzed in terms of time to resolve the emergency, by using analysis of videos by an independent rater. Group 1 resolved both crises significantly faster. HPS allowed us to detect design features that were common sources of error.

IMPLICATIONS: A human patient simulator may reduce the risk of learning-curve error in anesthesiologists using new, complex equipment without subjecting patients to risk. It can be used to highlight potential design errors in new equipment through usability testing or certify practitioners before they use new equipment.

*Department of Anaesthesia and Pain Manaement, Wellington Hospital, †National Patient Simulation Training Centre, Wellington Hospital, and ‡Wellington School of Medicine and Health Sciences, University of Otago, Wellington, New Zealand

Accepted for publication June 8, 2004.

Address correspondence and reprint requests to Brian Robinson, PhD, National Patient Simulation Training Centre, Wellington Hospital, Private Bag 7902, Wellington South, New Zealand. Address e-mail to brian.robinson@ccdhb.org.nz.

A number of new anesthesia delivery systems are being introduced into clinical practice. These systems have significant differences in circuit design, fresh gas flow (FGF) delivery, and ventilator control compared with common existing systems (1). Although many of these features may be useful, the effect of introducing different equipment on patient safety has not been thoroughly examined.

In anesthesia, patient safety depends on a complex interaction between a skilled practitioner and equipment (2–4). Equipment is involved in approximately 14%–30% of all intraoperative problems, and the anesthetic delivery equipment is the most common source of problems (5–7). Pure equipment failure is less common than equipment misuse. Human error is identified as a factor in up to 90% of equipment-related problems (6,8–12). Although anesthesia is becoming increasingly safe, critical incidents and adverse events still occur, and patients are harmed (10,11,13,14).

Poor equipment design and lack of familiarity with equipment are likely sources of latent errors and have been identified as major factors in many adverse events (2,8,9). New equipment designs can have unintended and detrimental effects on the interaction between practitioner and equipment and on patient safety (10,15).

In high-reliability industries, such as aviation, the introduction of new equipment involves rigorous testing, training, and certification. Equipment and the interface with the user is carefully studied and designed to reduce error and minimize the consequences of errors that do occur (3,4).

The typical introduction of new equipment in anesthetic practice involves an in-service training session by a company representative who demonstrates its use and supplies reference material. A representative may be present in the operating room to assist the practitioner during initial use of the equipment. This company employee is not likely to be an anesthesiologist.

We believe that this method of introducing new equipment into clinical practice is inadequate and exposes patients to increased risk. Human patient simulation (HPS) allows the interaction between practitioner and equipment to be put under stress without risk to patients. Latent errors and design faults, which may not be evident under normal circumstances, are more likely to manifest under these conditions (2,16–20). We postulated that simulating clinical use of novel complex anesthetic equipment should improve the ability to manage subsequent critical incidents and provide insight into potential design errors.

Back to Top | Article Outline

Methods

A randomized, controlled, prospective trial was designed with a Dräger Fabius GS Anesthesia Delivery Machine (Dräger, Telford, PA) and a METI HPS (METI, Sarasota, FL) to investigate our hypothesis. Our METI HPS is in a mock operating room, with operating room equipment and personnel. Patient monitoring (Datex AS3; Helsinki, Finland) and views of participants were recorded to videotape for subsequent analysis.

We approached all 18 anesthesiology trainees of the Wellington regional anesthesia training program for recruitment; none had previous experience with the Dräger Fabius GS. All had considerable experience with the HPS through involvement in departmental teaching programs. All had at least 1 yr of experience with the Datex-Ohmeda Excel SE anesthesia machines and with Ohmeda 7800 and 7900 ventilators (Ohmeda, Madison, WI). One trainee refused to participate, and two were unavailable over the study period. The 15 enrolled participants were given an information sheet explaining the purpose of the study and gave written, informed consent. The local Wellington Ethics Committee gave the study approval.

Participants were randomized to two groups by using a sealed-envelope technique. A stratified randomization was used to ensure that both groups had the same number of basic and advanced trainees.

All participants received a full in-service teaching session on the Dräger Fabius GS anesthetic-delivery machine from the same Dräger representative. This took 45–60 min. Participants attended singly or in pairs. The in-service teaching followed the format normally used by the representative before use of the machine in clinical practice.

Study participants were shown the novel features incorporated into the Dräger Fabius GS anesthetic machine. These include electronic gas flow monitoring, automated leak and compliance checks, an electrically driven ventilator with a noncollapsing bellows system capable of entraining room air during insufficient or absent FGF, a ventilator that will function without any FGF, a common gas outlet (CGO) controlled by a small lever, vaporizer selection with a lever system, a novel breathing-circuit configuration, an airway pressure-limiting (APL) valve that resembles a positive end-expiratory pressure valve on existing machines at Wellington Hospital, and electronic switching from ventilator to manual ventilation.

Participants randomized to Group 1 also simulated clinical use of the Dräger machine with the HPS. They were encouraged to formally check the machine before using it in a straightforward simulated clinical scenario. This consisted of anesthetizing a healthy patient for a minor orthopedic procedure in the presence of the principle investigator, who ensured that they were able to use the machine to perform the following actions: administer oxygen to the patient, mask-ventilate, use both volume-controlled intermittent positive-pressure ventilation (IPPV) and pressure-controlled ventilation, administer volatile anesthesia, change ventilator control, and return the patient to manual and spontaneous ventilation at case end. Any questions regarding the anesthesia delivery system were answered by the principal investigator (PD). Group 2 did not take part in the simulation-based teaching session.

All participants completed a questionnaire focusing on their perceptions of using this machine in clinical practice (Table 1). Responses to these questions were both by a rating scale (1–5) and written comments.

Table 1

Table 1

After completing the questionnaire, all 15 registrants participated in 2 clinical scenarios simulated with the HPS. The order of the scenarios was randomized, and both scenarios were complicated by predefined problems.

Back to Top | Article Outline

Scenario A

This scenario tested the participant’s understanding of the pressure-limiting function (Pmax) that operates during volume-controlled IPPV. A young asthmatic woman required anesthesia for laparoscopy. A full description of the case was provided. The patient was on the operating table with monitoring attached and drugs drawn up. The ventilator was set to volume control with a Pmax of 15 cm H2O. After the induction of anesthesia and tracheal intubation, the simulated patient developed severe bronchospasm. Monitoring demonstrated reduced pulmonary compliance, an abnormal capnogram, and hypoxia. As airway pressure increased and reached the Pmax of 15 cm H2O, the delivered tidal volume decreased to less than the preset volume of 500 mL. Clinical examination demonstrated poor chest wall excursion, bilateral wheeze, and high airway pressure, with manual ventilation at normal tidal volumes. The times from intubation to the recognition of severe bronchospasm and initiation of effective treatment were recorded.

Back to Top | Article Outline

Scenario B

This scenario tested the participant’s understanding of the FGF control and the ability of the Dräger Fabius GS to ventilate the patient with entrained room air. A young man required a computed tomography (CT) brain scan for a severe head injury. He had pulmonary contusions, resulting in a high intrapulmonary shunt and poor pulmonary compliance. A full history, examination, and investigations were provided. Endotracheal tube and invasive monitoring were in place. The patient had just been placed on the CT scanning bed, and the monitoring was attached to the anesthetic machine. The intensive care unit (ICU) nurse was manually ventilating the patient by self-inflating resuscitation bag. Pmax was set at 15 cm H2O; a Mapleson F circuit had been used for the previous case and was still connected to the CGO. The anesthesiologist handing over the case removed this in the presence of the study participant but failed to return the FGF to the circle breathing system. Other ventilator settings were as for the patient while in the ICU.

The FGF on the Dräger Fabius is easily switched between the circle system and an external circuit by a small lever on the side of the machine. When the FGF is being delivered to an external source or, in this case, the operating room, the ventilator will still function and deliver tidal volume by the entrainment of room air into the circle system. If the position of the external/circle FGF control was not corrected, the HPS rapidly became hypoxic when ventilated with the Dräger machine. The HPS also became tachycardic and the blood pressure increased if no volatile anesthetic was delivered. With this patient’s poor pulmonary compliance, an adequate tidal volume was not delivered until the Pmax setting was altered or the ventilatory mode was changed to Pressure Control, with appropriate settings.

The times were recorded from completion of the handover of care to the provision of an adequate fraction of inspired oxygen, minute volume, and anesthetic vapor. Assessments of the times taken to crisis resolution were by videotape analysis by an independent observer (CC) blinded to group allocation. We also recorded errors made by study participants during the management of the two clinical crises from direct observation and independent review of the videotapes by the principal investigator and the investigator blinded to group allocation (CC).

Times to crisis resolution were analyzed with two-tailed Student’s t-test, with unequal variances. Statistical analysis was performed with Stata statistical software, Release 7.0 (StataCorp).

Back to Top | Article Outline

Results

There were no statistically significant differences between groups for any of the question scores (Table 1). In the written comments, three participants said that they would feel confident using the machine after a day of using it or after a period of familiarization. One participant felt confident to use the machine in a normal case, but not in a crisis. Four participants did not feel confident to use the machine, either because they had not used it on patients or because it did not feel familiar. Asked whether this new machine would distract them from the patient, five believed that it would because of a lack of familiarity; one believed that reflex responses would take longer and would need to be double-checked, which would distract from the patient. Four participants believed that using the new machine could lead to errors; four believed that the opportunity to familiarize themselves with the equipment or to use it in a clinical simulation would reduce the chance of making errors, and one person felt unlikely to make errors with the new equipment.

Participants from both groups noted many differences between the new machine and existing equipment. Nine participants described differences in the operation of the bellows or their ability to entrain room air, seven noted the differences in APL valve design, four noted the different routing of FGF through the circuit, and five noted the different ventilator display or controls. When tested in the simulated clinical scenarios, participants in Group 1, who had received the simulator training, instituted appropriate treatment significantly faster in both clinical scenarios (Table 2).

Table 2

Table 2

In Scenario A, no errors were made by Group 1 participants, whereas 14 errors were made by Group 2 participants. Of these, three participants had difficulty with or were unable to correctly work the APL valve, and one was unable to manually ventilate the patient because of this. Five participants had difficulty in turning the ventilator on or switching back to manual ventilation. Three participants misdiagnosed bronchospasm as a machine problem. Two participants were unable to use the desired vaporizer.

In Scenario B, three errors were made by members of Group 1 and eight by members of Group 2. The errors made by Group 1 were an inability to turn on the sevoflurane vaporizer, wrongly diagnosing a circuit leak, and difficulty changing from ventilator to manual hand ventilation. In Group 2, three participants were unable to turn on the ventilator or return the patient to manual ventilation. Two participants were unable to make desired changes to ventilator settings. Two participants were unable to use the selected vaporizer. One participant thought that the problem was pipeline oxygen failure. The ability of the ventilator to function without FGF, the circuit/external CGO controller, and the ventilator’s ability to entrain room air had been commented on in the training period by nine participants. However, all but one participant (Group 1) initially ventilated the patient with room air.

Back to Top | Article Outline

Discussion

After the standard in-service teaching, many study participants had difficulty using the anesthetic machine in a clinical scenario. For example, although participants had noted that the ventilator would entrain room air if there was no FGF, once placed in a stressful situation they failed to use this knowledge and, as a result, ventilated the patient with room air. Standard teaching and introduction of new equipment may be adequate under routine clinical conditions, where practitioners have time to work through problems from first principles. However, this can be laborious and may result in an increased risk of error. It would be preferable to develop practical familiarity with complex new equipment in a safe environment. If the practitioner encounters a clinical emergency when using unfamiliar equipment, patient safety may depend on rapid problem recognition, and the risk of serious error will increase. The initial presence of a company representative is unlikely to provide a robust defense against these risks.

We demonstrated no correlation between study participants’ confidence in using the new equipment, as assessed in the questionnaire, and their subsequent performance in the clinical scenarios. Some participants rated their confidence high or believed that they had a small likelihood of making errors but later made serious errors with the new equipment. Some participants noted important design features of the new machine, but then failed to use these features appropriately. We noted no statistically significant difference between groups for any of the question scores, despite the significantly superior performance of Group 1 participants. This strongly suggests that after introduction to new and complex equipment, anesthesiologists cannot reliably assess their ability to safely use the equipment in clinical practice.

HPS provides an excellent opportunity to assess equipment design (13). The equipment and the interaction between practitioner and equipment can be tested repeatedly under different conditions, without exposing patients to potential risk. The machine-user interface can also be put “under stress” in simulated crises, which may uncover latent design errors and deficiencies in training. Rigorous testing with HPS during the design process could improve equipment design. The ability of the Dräger Fabius GS to allow unintended patient ventilation with room air appeared as a latent design fault and a significant source of error in Scenario B.

Anesthesia delivery equipment is becoming increasingly complex (1). As the complexity of the system increases, so does the likelihood for error (2). New equipment should be designed both to reduce the likelihood of errors and to increase the early detection of errors that do occur, to stop the progression to patient harm. The HPS provides a means to investigate the safety of equipment design in a clinical emergency by producing a situation in which the risk of errors is increased.

Our study demonstrates that, after the standard introduction to complex and unfamiliar anesthesia equipment, practitioners are unable to self-assess their competence to use that equipment. They are also likely to make multiple errors, which may interact with latent design faults to produce critical incidents. We have demonstrated that HPS can improve the ability of practitioners to safely use new equipment. Although other methods of hands-on instruction may also be useful, we have demonstrated that HPS is an effective training device to ensure that practitioners are competent to use new equipment in both straightforward and crisis situations before they use the equipment with real patients.

Back to Top | Article Outline

References

1. Sloan IA. The new anesthetic machines. Clin J Anaesth 2002;47:201–4.
2. Gaba DM, Fish KJ, Howard SK. Crisis management in anesthesiology. Philadelphia: Churchill Livingstone, 1994.
3. Leape LL. Error in medicine. JAMA 1994;272:1851–7.
4. Reason J. Human error: models and management. BMJ 2000;320:768–70.
5. Fasting S, Gisvold SE. Equipment problems during anaesthesia: are they a quality problem? Br J Anaesth 2002;89:825–31.
6. Cooper JB, Newbower RS, Long CD, McPeek B. Preventable anesthesia mishaps: a study of human factors. Anesthesiology 1978;49:399–406.
7. Webb RK, Russell WJ, Klepper I, Runciman WB. Equipment failure: an analysis of 2000 incident reports. Anaesth Intensive Care 1993;21:673–7.
8. Runciman WB, Sellen A, Webb RK, et al. Errors, incidents and accidents in anaesthetic practice. Anaesth Intensive Care 1993;21:506–19.
9. Cooper JB, Newbower RS, Kitz RJ. An analysis of major errors and equipment failures in anesthesia management: considerations for prevention and detection. Anesthesiology 1984;60:34–42.
10. Weinger MB. Anesthesia equipment and human error. J Clin Monit 1999;15:319–23.
11. Short TG, O’Regan A, Lew J, Oh TE. Critical incident reporting in an anaesthetic department quality assurance programme. Anaesthesia 1992;47:3–7.
12. Williamson JA, Webb RK, Sellen A, et al. Human failure: an analysis of 2000 incident reports. Anaesth Intensive Care 1993;21:678–83.
13. Merry AF, Webster CS, Weller J, et al. Evaluation in an anaesthetic simulator of a prototype of a new drug administration system designed to reduce error. Anaesthesia 2002;57:256–63.
14. Caplan RA, Vistica MF, Posner KL, Cheney FW. Adverse anesthetic outcomes arising from gas delivery equipment: a closed claims analysis. Anesthesiology 1997;87:741–8.
15. Tenner E. Why things bite back: technology and the revenge of unintended consequences. New York: Vintage Books, 1997.
16. Forrest FC, Taylor MA, Postlethwaite K, Aspinall R. Use of a high-fidelity simulator to develop testing of the technical performance of novice anaesthetists. Br J Anaesth 2002;88:338–44.
17. Schwid HA, Rooke GA, Carline J, et al. Evaluation of anesthesia residents using mannequin-based simulation. Anesthesiology 2002;97:1434–44.
18. Devitt JH, Kurrek MM, Cohen MM, et al. Testing internal consistency and construct validity during evaluation of performance in a patient simulator. Anesth Analg 1998;86:1160–4.
19. Devitt JH, Kurrek MM, Cohen MM, et al. Testing the raters: inter-rater reliability of standardised anaesthesia simulator performance. Clin J Anaesth 1997;44:924–8.
20. Weller JM, Bloch M, Young S, et al. Evaluation of high fidelity patient simulator in assessment of performance of anaesthetists. Br J Anaesth 2003;90:43–7.
© 2004 International Anesthesia Research Society